{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:LSUUMCDBDIGAPAERFAH4OXXD4D","short_pith_number":"pith:LSUUMCDB","schema_version":"1.0","canonical_sha256":"5ca94608611a0c078091280fc75ee3e0c9c67567546cef9a8c66d52fd3d71c73","source":{"kind":"arxiv","id":"2306.09478","version":2},"attestation_state":"computed","paper":{"title":"Understanding and Mitigating Extrapolation Failures in Physics-Informed Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Luca D'Amico-Wong, Lukas Fesser, Richard Qiu","submitted_at":"2023-06-15T20:08:42Z","abstract_excerpt":"Physics-informed Neural Networks (PINNs) have recently gained popularity due to their effective approximation of partial differential equations (PDEs) using deep neural networks (DNNs). However, their out of domain behavior is not well understood, with previous work speculating that the presence of high frequency components in the solution function might be to blame for poor extrapolation performance. In this paper, we study the extrapolation behavior of PINNs on a representative set of PDEs of different types, including high-dimensional PDEs. We find that failure to extrapolate is not caused "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2306.09478","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-15T20:08:42Z","cross_cats_sorted":[],"title_canon_sha256":"be4c400e921bca587e0c0380ed0f944444b4356a6c352c8defbab911c6aa016c","abstract_canon_sha256":"b176f4480647369e31dd3f14fa450f1306a783283370dd2caa356eb3dd7a2831"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:16:43.007889Z","signature_b64":"dSbmDZ4o1HheCAOgCNEDX9Wg+JUQg3kUuvt83cPR27Q/hJdjq1Dk1I8OxvaMILK3Gith5uGhNl2NH7EAJGvRBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5ca94608611a0c078091280fc75ee3e0c9c67567546cef9a8c66d52fd3d71c73","last_reissued_at":"2026-07-05T07:16:43.007425Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:16:43.007425Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding and Mitigating Extrapolation Failures in Physics-Informed Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Luca D'Amico-Wong, Lukas Fesser, Richard Qiu","submitted_at":"2023-06-15T20:08:42Z","abstract_excerpt":"Physics-informed Neural Networks (PINNs) have recently gained popularity due to their effective approximation of partial differential equations (PDEs) using deep neural networks (DNNs). However, their out of domain behavior is not well understood, with previous work speculating that the presence of high frequency components in the solution function might be to blame for poor extrapolation performance. In this paper, we study the extrapolation behavior of PINNs on a representative set of PDEs of different types, including high-dimensional PDEs. We find that failure to extrapolate is not caused "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.09478","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2306.09478/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2306.09478","created_at":"2026-07-05T07:16:43.007483+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.09478v2","created_at":"2026-07-05T07:16:43.007483+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.09478","created_at":"2026-07-05T07:16:43.007483+00:00"},{"alias_kind":"pith_short_12","alias_value":"LSUUMCDBDIGA","created_at":"2026-07-05T07:16:43.007483+00:00"},{"alias_kind":"pith_short_16","alias_value":"LSUUMCDBDIGAPAER","created_at":"2026-07-05T07:16:43.007483+00:00"},{"alias_kind":"pith_short_8","alias_value":"LSUUMCDB","created_at":"2026-07-05T07:16:43.007483+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.04000","citing_title":"SPLIT-PINN: Separable Probability Learning Technique via Physics-Informed Neural Networks for High-Dimensional Probabilistic Modeling","ref_index":86,"is_internal_anchor":false},{"citing_arxiv_id":"2501.12281","citing_title":"MoGERNN: An Inductive Traffic Predictor for Unobserved Locations","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06561","citing_title":"Accelerating 4D Hyperspectral Imaging through Physics-Informed Neural Representation and Adaptive Sampling","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LSUUMCDBDIGAPAERFAH4OXXD4D","json":"https://pith.science/pith/LSUUMCDBDIGAPAERFAH4OXXD4D.json","graph_json":"https://pith.science/api/pith-number/LSUUMCDBDIGAPAERFAH4OXXD4D/graph.json","events_json":"https://pith.science/api/pith-number/LSUUMCDBDIGAPAERFAH4OXXD4D/events.json","paper":"https://pith.science/paper/LSUUMCDB"},"agent_actions":{"view_html":"https://pith.science/pith/LSUUMCDBDIGAPAERFAH4OXXD4D","download_json":"https://pith.science/pith/LSUUMCDBDIGAPAERFAH4OXXD4D.json","view_paper":"https://pith.science/paper/LSUUMCDB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.09478&json=true","fetch_graph":"https://pith.science/api/pith-number/LSUUMCDBDIGAPAERFAH4OXXD4D/graph.json","fetch_events":"https://pith.science/api/pith-number/LSUUMCDBDIGAPAERFAH4OXXD4D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LSUUMCDBDIGAPAERFAH4OXXD4D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LSUUMCDBDIGAPAERFAH4OXXD4D/action/storage_attestation","attest_author":"https://pith.science/pith/LSUUMCDBDIGAPAERFAH4OXXD4D/action/author_attestation","sign_citation":"https://pith.science/pith/LSUUMCDBDIGAPAERFAH4OXXD4D/action/citation_signature","submit_replication":"https://pith.science/pith/LSUUMCDBDIGAPAERFAH4OXXD4D/action/replication_record"}},"created_at":"2026-07-05T07:16:43.007483+00:00","updated_at":"2026-07-05T07:16:43.007483+00:00"}